Kobayashi Yukihiro, Yokose Tomoyuki, Kawamura Kimihiko, Iwasaki Seiji, Murata Yukinori, Onuma Shinkichi, Hasebe Takahiro, Nagai Kanji, Sasaki Satoshi, Ochiai Atsushi
Clinical Laboratory Division, National Cancer Center Hospital East, Chiba, Japan.
Cancer. 2005 Feb 25;105(1):44-51. doi: 10.1002/cncr.20686.
Recently, peripheral lung adenocarcinomas (PLA) measuring < or = 3 cm in greatest dimension often have been diagnosed using diagnostic radiology. The objective of the current study was to determine which cytologic factors are associated with a favorable outcome and an unfavorable outcome in patients with PLA.
Imprint smears from 134 patients with PLA were examined. Sixteen cytologic factors, including necrosis, cellular distribution, overlapping of cell clusters, cluster aggregation, cluster size, cluster thickness, nuclear irregularity, nuclear size, variation in nuclear size, multinucleated cells, intranuclear inclusions, type of intranuclear inclusions, appearance of nucleoli, eosinophilic nucleoli, multinucleoli, and mitosis, were evaluated using univariate and multivariate analyses. A counting method was used to determine the prognosis for individual patients.
In the univariate analysis, a cluster size that measured > or = 831 microm in short dimension (P = 0.0011), moderate or severe nuclear irregularity (P = 0.0030), > or = 5 multinucleated cells per 100 tumor cells (P = 0.0047), moderate or severe variation in nuclear size (P = 0.0061), medium or large nuclear size (P = 0.0169), and > or = 1 mitotic cell per 100 tumor cells (P = 0.0412) were associated significantly with a poor outcome. In the multivariate analysis, cluster size in short dimension (P = 0.0018), multinucleated cells (P = 0.0066), and nuclear irregularity (P = 0.0310) were found to be independent prognostic factors.
The combination of cytologic features using intraoperative imprint smears, namely, cluster sizes < or = 830 microm in short dimension, < or = 4 multinucleated cells per 100 tumor cells, and mild nuclear irregularity, may provide favorable predictive information in patients with PLA.
